Want to Buy the Next Google or Apple? Let's See if You Can.
Imagine if you could have gotten in on Google’s IPO, or if you had recognized the brilliant future that Apple’s iMac and iPod signified. These are the types of investment calls that make investors salivate. But were people who got in early making well-reasoned decisions, or were they just lucky?
YCharts is constantly seeking to outperform the market. We would love to claim that we have a reliable way to locate the future Apples and Googles of the world before they grow, but the successful strategies that we uncovered do not frequently suggest purchasing high-growth companies.
We wanted to know why finding future growth was so difficult, so we read the research and extended it with our own analysis. The results were both were surprising and invaluable. So we decided to share them with you.
From August of 2004 to December of 2010, Google’s (GOOG) price increased from $100 to $600 per share. And Apple (AAPL), from the end of 2000 to the end of 2010 returned 43.35X. Those are returns that put dollar signs in any investors’ eyes.
Some still kick themselves for not investing in these companies before they grew. Now what happened looks obvious: Google was the search engine that the tech guys were using. Steve Jobs had returned to Apple and was sowing the seeds of success.
But we can hear the skeptics already. Yahoo! was competing heavily with Google then; Google might have lost. Who could have known that Apple was going to win with its pretty machines? Point taken.
So fast-forward to the end of 2006. Google had grown earnings by more than 120% annualized for 3 years. Apple had grown by more than 30% per year, sales were growing similarly, and prices were increasing.
Three obvious facts were right in front of investors: growth, growth and growth! The paths seemed clear, and history shows us there were still profits to be earned: Apple was still about ¼ of its end of 2010 value and Google had another 30% to rise. Certainly we could have predicted with 3+ years of historical growth to support.
Any skeptics still there?
YCharts searched the world of investing knowledge and found three very smart ones: Professors Louis Chan, Jason Karceski and Joseph Lakonishok. These three put growth to the test in a 2003 paper.
Before getting back to Apple and Google, a short detour in the form of a bet.
Imagine that a man walks up to you and tells you this: He will pick a public company at random from a very select group. Every company in the group grew earnings(1) faster than 50% of all other companies not just last year, but for the past 3 years. Like Apple and Google, they grew, and they did it consistently.
The man offers you $10,000 if the company’s earnings grow in the top 50% relative to all U.S. companies again this year. But if the company’s growth rate is in the bottom 50%, you must pay him $11,000. Do you take the bet?
The bet is extraordinarily tempting. In fact, most investors we know look at past earnings growth in order to predict future earnings growth.
Unfortunately, if you take the bet, the odds are against you. There is no evidence that past earnings growth predicts future earnings growth. In fact, if we look at growth rates of historically fast-growing firms over the next 5 years, the results are astounding.
After one year, only 48% of historically fast-growing firms continue to grow faster than the median firm. After two years, only 23.7% beat the median for two years, and so on. The full results are summarized below:
Put simply, this table shows us that trying to pick a stock’s future growth path based on past growth is like trying to guess if a coin will come up heads or tails when you know that the last toss was a heads. The previous toss tells you nothing.
However, many investors still fall for the historical growth fallacy. But the error is very similar to another problem with which most people are aware: the investing newsletter scam.
Each month, a scam artist sends 1000 newsletters to 1000 different people. 500 letters predict that the market will increase during the month, and 500 predict a decline. At the end of the month, half of the people received “correct” letters with certainty, and the scam artist sends 500 more letters to them. 250 predict an increase in the market, and 250 the contrary. Again, half will be correct. Repeat for 6 months, and about 15 people receive letters which were correct every month.
The unlucky 15 who received all correct predictions often reason that the past correct newsletters prove that future ones will be correct. Obviously they’re incorrect.
Similarly, investors tend to focus on companies that grew historically, citing past growth as proof that the company should grow more in the future. But the broader view shows that these companies success can be explained by luck - the result of a string of favorable coin flips.
When viewed through this lens, betting on future growth looks questionable. Going back to the situation in 2006, Apple and Google’s continued growth at above median rates is far from obvious.
There is one other issue to consider: one year of 100% growth is a larger move than 5 years of 14% growth (14% growth is higher than the median 1 year growth rate, 1951-1997). Therefore, companies do not have to grow quickly year after year as long as they have one or two good years. To control for this fact, we looked at long-term compounded growth in earnings(2) in order to see how companies grow over extended periods of time.
Now lets see how likely it is that we could have picked Apple or Google before their growth kicked in.
Apple’s earnings grew at an annualized rate of 91%(3) over the five year period ending in December 2010, which puts it well above the 95th percentile for growth. Not impossible to find, but certainly improbable.
To add perspective: If you only select companies from the S&P 500 there are only 25 stocks that are above the 95th percentile for growth - you cannot fill a 35 stock portfolio with them.
Google managed to grow earnings at over 95% annualized for the 5 year period ending in December of 2008, placing it in a similar group. Again, finding the company before it hit its growth streak seems difficult.
Take one more example to drive home the abstract statistics above: Dell. From the beginning of 1991 to the beginning of 2000, Dell’s price increased more than 600X and earnings per share increased by 41.9X (or 51% annualized for 9 years). At that point, it was looking like Apple is now. But by 2001, growth hopes had faded as earnings per share fell 25% from 2000-2002. 12 years later, prices have still not returned to 1999 levels.
The market was betting on growth, and growth was unpredictable. Since 2002, Dell has had 5 years of positive earnings growth and 3 years of negative earnings growth - which is hardly a continuation of the trend of the 1990s.
Occasionally, people ask us why YCharts Pro doesn’t often recommend growth stocks or develop strategies with growth. Believe us: we are searching. We read the research and run our own tests. We just haven’t found evidence that we can identify companies before they hit their growth spurts.
We think that it is nearly impossible to search through hundreds or thousands of stocks to locate the Apples and Googles of 2000-2010 while simultaneously avoiding the Dells of 2000-2002. Therefore, when we read about strategies that rely on the future growth of companies to succeed, we keep an open mind, but we remain skeptical.
(1): We use “earnings” as shorthand for earnings before extraordinary items.
(2): The data is Compustat Data from 1985-2010 in 5 separate 5-year increments. There are no overlapping increments for 5 year growth rates, but 10 year rates will have overlap.
(3): Earnings before extraordinary items grew at 91%. Diluted EPS including extraordinary items grew at only 57%.
We are by no means arguing that Apple and Google will fail to grow in the future. They might. However, we do not believe that investors can predict the companies’ growth accurately before the fact.
We would be delighted to hear about any research that contradicts the findings in this article if it is out there, specifically related to the predictability of growth rates.
This article does bypass the question of Momentum strategies (buying based on recent price appreciation), for which there is some evidence, but we view that is a separate strategy from growth.